PENELOPIE: Enabling Open Information Extraction for the Greek Language through Machine Translation

In this paper we present our submission for the EACL 2021 SRW; a methodology\nthat aims at bridging the gap between high and low-resource languages in the\ncontext of Open Information Extraction, showcasing it on the Greek language.\nThe goals of this paper are twofold: First, we build Neural Machine Translation\n(NMT) models for English-to-Greek and Greek-to-English based on the Transformer\narchitecture. Second, we leverage these NMT models to produce English\ntranslations of Greek text as input for our NLP pipeline, to which we apply a\nseries of pre-processing and triple extraction tasks. Finally, we\nback-translate the extracted triples to Greek. We conduct an evaluation of both\nour NMT and OIE methods on benchmark datasets and demonstrate that our approach\noutperforms the current state-of-the-art for the Greek natural language.\n